A high-performance security auditing tool designed to scan AI agent code and skill sets for malicious patterns and vulnerabilities.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-security-skill-scanner-gitee
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install agent-security-skill-scanner-gitee using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Agent Security Scanner?
The Agent Security Scanner is a comprehensive auditing tool specifically built to ensure the safety of Openclaw Skills and agent-based workflows. It utilizes over 183 detection rules to identify malicious code across popular languages such as Python, JavaScript, Go, and Shell.
By integrating this scanner into your development pipeline, you can automatically verify the integrity of external dependencies, local files, and third-party repositories. It bridges the gap between AI automation and cybersecurity, providing developers with a robust mechanism to evaluate the risk levels of scripts before execution.
Agent Security Scanner Use Cases
- Identifying malicious code patterns in newly downloaded Openclaw Skills.
- Auditing local Python, JavaScript, or Shell scripts for common security vulnerabilities.
- Verifying the safety of NPM packages or GitHub repositories before integration.
- Automated security scanning of deployment configurations and YAML files.
How Agent Security Scanner Works
- The user triggers the scan via the asc-scan CLI tool or a scheduled cron job.
- The scanner automatically identifies the target type, whether it is a local file, a directory, or a remote repository.
- Static analysis is performed against a library of 183+ detection rules covering over 10 attack types.
- Optional LLM-powered deep analysis is invoked if enabled to provide additional context for complex vulnerabilities.
- A risk score from 0 to 100 is generated, categorizing the findings into low, medium, or high risk levels.
- The tool generates a report in the desired format (Standard, Verbose, or JSON) for manual review or automated processing.
Agent Security Scanner Setup
To get started with this security tool for your Openclaw Skills, follow these steps:
# Clone the repository
git clone https://github.com/caidongyun/agent-security-skill-scanner.git
cd agent-security-skill-scanner/release/v5.1.0
# Install the CLI tool
chmod +x asc-scan
sudo ln -sf $(pwd)/asc-scan /usr/local/bin/asc-scan
# Verify installation
asc-scan --version
Agent Security Scanner Data Schema & Taxonomy
The scanner organizes findings based on a severity taxonomy and supports structured data exports:
| Attribute | Type | Description |
|---|---|---|
| risk_score | Integer | A value from 0-100 indicating the threat level. |
| risk_level | String | Categorized as Low (0-19), Medium (20-49), or High (50-100). |
| language | String | The detected source language (Python, JS, etc.). |
| rules_triggered | Array | List of specific security rules violated by the code. |
| output_format | Format | Supports default text, verbose logging, and JSON objects. |
Agent Security Scanner Advanced Features
- Intelligent LLM-powered analysis for deep-diving into suspicious code logic.
- Background daemon mode for continuous monitoring of directory changes in Openclaw Skills.
- Flexible whitelist mechanism to reduce false positives in known safe environments.
- Multi-channel notification support including Feishu Webhooks and Email alerts.
- Seamless integration with crontab for scheduled enterprise-grade security audits.
SKILL.md
Loading
Related Openclaw Skills
Caption Generator in Video
An AI-powered skill for Openclaw Skills that automatically generates high-quality captions and subtitles for video content using cloud-based GPU rendering.
Zhang Xuefeng Perspective: Pragmatic Career & Education Strategy
A pragmatic AI thinking framework that applies Zhang Xuefeng's high-certainty decision models to education and career planning.
x-search
A powerful search tool that leverages the xAI Grok API to find, summarize, and cite real-time posts from X (Twitter).
Maton Browse Plan
A Chrome extension and agent skill that analyzes browsing activity to proactively suggest and configure Maton API OAuth connectors.
FlyAI Hotel Search & Compare
A powerful hotel discovery and comparison tool for AI agents that filters stays by landmarks, property types, and nightly budget in CNY.
gh-skillscan
A security auditing tool designed to scan Openclaw Skills for potential threats and malicious patterns before installation.
Featured*
